MAGNETIC COMPASS

DEVIATION AND
CORRECTION

A MANUAL OF THE THEORY OF THE DEVIATIONS

AND MECHANICAL CORRECTION OF MAGNETIC
COMPASSES IN SHIPS

BY
W. DENNE. Extra Master
F.Inst.Nav., Assoc. R.I.N.A.

REVISED BY

Captain A. N. COCKCROFT

GLASGOW
BROWN, SON & FERGUSON, LTD.,
NAUTICAL PUBLISHERS
52 DARNLEY STREET
Copyright in all countries signatory to the Berne Convention
All rights reserved

TO MY FRIEND AND WAR-TIME SHIPMATE

LIEUTENANT-COMMANDERHAROLD W. LARSEN,
PH.D., R.N.V.R.

First Edition 1951

Second Edition 1968
Third Edition 1979

ISBN 0 85174 332 3

1979 BROWN, SON & FERGUSON, LTD., GLASGOW, G41 2SG
Printed and Made in Great Britain
 PREFACE

THIS work, in so far as the mathematical theory of deviations

is concerned, is based entirely on the Admiralty publication The
Theory of the Deviations of the Magnetic Compass prepared by the
Admiralty Compass Observatory, Slough, a fact which is gratefully
acknowledged.
The book has been written in an endeavour to explain as simply
as possible all that is involved in the correction, adjustment and
maintenance of magnetic compasses on board ships. With the
exception of the Admiralty Manualfor the Deviations of the Compass,
now out of print, and the above mentioned work written for the
Royal Navy, no other book has been published giving the full
mathematical basis on which the practical application is based.
The preface to the Admiralty Manual stated that the theory
was intended only for the skilled mathematician. It would appear
that this rather frightening comment has hitherto discouraged the
majority of seamen, compass adjusters and others from any en-
deavour to obtain a thorough grasp of the subject, consequently
much has had to be learned by rule of thumb.
The fact is, however, that the theory as presented in the Admiralty
Manual is capable of being understood only by a skilled mathe-
matician. The actual mathematics is comparatively simple and
involves no more knowledge than that contained in the first few
chapters of a book on elementary trigonometry together with a
knowledge of algebra up to and including the use of simple equations.
The methods used in the mechanical correction of the compass
are based on theory which is put into practical form only through
the manipulation of three fundamental equations. The theory is
derived from the fundamental laws relating to magnetic fields and
materials. The mathematical manipulation gives the magnetic forces
involved in terms of the deviation of the compass which they
cause and the Exact Coefficients for any magnetic heading of the
ship, but it will not give the deviation explicitly in terms of the
Exact Coefficients and the Compass Course, which is what is
generally required.
By making certain assumptions and using successive approxima-
tions a Fourier's series is obtained which gives the deviation in
terms of approximate coefficients and the Compass Course. It also

PREFACE
gives the relationships between the exact and approximate co-
efficients. The accuracy depends on how many of these approximate
coefficients are used. Normally only the first five are considered
on the assumption that the compass has been suitably placed in
relation to the magnetic fields in the ship. By using the approximate
coefficients found by analysis, correctly or approximately, usually
approximately, or by methods based on approximate analysis, the
compass is initially corrected and later adjusted from time to time
as changes in the sub-permanent forces in the ship, or other con-
tingencies, may require. The Heeling Error involves even more
approximations.
It is possible, therefore, that a blind use ofthe first five approximate
coefficients for the correction or adjustment of the compass may
lead to a false sense of security and perhaps considerable error.
This might be especially true in present day ships with so many
modern navigational aids of an electrical nature fitted in the vicinity
of the compass.
One must also bear in mind that until the ship's electricity
supply is made infallible the magnetic compass will remain one
of the most important and reliable of all navigational aids.
It is felt that though the seaman in general may not be a skilled
mathematician, it is unreasonable to assume that he has no
knowledge of mathematics nor that he and others concerned have
no interest in the subject when it is required to enhance their
professional knowledge. One critic, however, has suggested to the
author that the book may be studied very profitably in its present
form without a full understanding of the mathematics providing
the necessity for its inclusion is understood.
The heart of the book, Chapters VIII to XIV inclusive, deals
with the magnetism of the ship, the evaluation of the coefficients
and the heeling error. The magnetism is dealt with under three
separate chapter headings-preliminary, permanent magnetism, and
induced magnetism. The latter chapter explains the conception of
the nine rods and the application of their "signs". Here the author
has indulged in a certain amount of repetition of description which
may offend the pedant but it has been done deliberately to drive
the conception home.
Chapter XI explains in detail the method of obtaining the expres-
sion involving the exact coefficients. The full mathematical process
including all the "steps" is given separately in the next chapter,
allowing the reader to omit or reserve the bulk of the mathematics
for separate study, if he so desires, without losing any of the
PREFACE
arguments leading to the final result. Similarly, the full mathematical
process involved in deriving the heeling error expression is kept to
the end of that chapter.
The subject matter of subsequent chapters dealing with analysis,
theory of mechanical correction and errors of the deflector, in
some instances, does involve more advanced mathematics and only
the results of these premises have been given. Should it be found
in the light of experience that a fuller exposition of these subsidiary
subjects would serve a useful purpose, it can be included in a later
edition. It has been the aim of the author to include only that
which is essential to impart a complete understanding of the main
subject, nevertheless it will be found that some hitherto unpublished
information is included.
Chapters IV to VII deal with magnetism in general, fundamental
facts, the earth's magnetism and magnetic measurements. These
chapters are merely intended to form a collection of facts and laws
which are required in order to understand the effects of the earth's
and the ship's magnetism on the magnetic compass and how the
various forces may be measured. They in no way constitute a
complete work on the subject. There are many excellent textbooks
which include this section of physics, and to these the reader is
referred if he requires greater detail. The methods of approach to
the subject used in these four chapters are similar to those used in
Text-book of Physics, by Duncan and Starling, Intermediate Physics,
by R. A. Houston, and Magnetism and Electricity, by S. G. Starling,
a fact which is hereby acknowledged.
Chapters II and III are included as a form of revision of the
subjects of vectors and trigonometrical ratios. Chapter I is purely
introductory giving some advice as to the use of the book with a
few remarks on the transposition of terms in algebraic equations.
Some worked examples of problems have been included at the end
of the book.
My grateful thanks are due to Captain J. H. Quick for his most
helpful criticisms and suggestions. to Mr Maurice Disney of the
Honourable Company of Master Mariners for editing the book, to
Captain O. Fletcher for his criticisms of the mathematics, to the
Director and Officers of the Admiralty Compass Observatory from
whom the author has obtained, from time to time, much valuable
information, to the Ministry of Transport for permission to publish
the book and to the Director of the Meteorological Office for
permission to include the analysis of the deviations of the S.S.
Weather Recorder. In addition I would like to record my apprecia-
"
 PREFACE

tion of the help given by my wife and also by Captain A. N. Manson

with the proof reading of the book.
It should be emphasised that any faults that may be found are
entirely due to the author, who would be very grateful for informa-
tion concerning any errors or omissions that may still remain. PREF ACE TO THE SECOND EDITION
WATFORD 1950.
IN this revised edition one or two obvious errors which had crept
into the text of the former edition have been rectified and some
minor modifications made to the original text. In response to many
requests the author has added the proof of the "slewing of the
spheres" and also a section on "pitching error".
EDINBURGH, 1967.

PREFACE TO THE THIRD EDITION

ALL c.g.s. units have been converted to S.l. units and various
corrections have been made to the text. Some examples relating
only to general magnetism have been withdrawn.
 CONTENTS

CHAP. PAGE

PREFACE VB

DEFINITIONS OF TERMS XV

T ABLE OF SYMBOLS AND THEIR MEANINGS XVlll

T ABLE OF FORMULAE AND LAWS. xx

I. INTRODUCTION 1
II. SPACE AND VECTOR DIAGRAMS. Parallelogram and
Triangle of Forces 6
III. TRIGONOMETRICAL RATIOS 11
IV. MAGNETISM. General . 19
V. FUNDAMENTAL FACTS OF MAGNETISM 22
VI. THE EARTH'S MAGNETISM 36
VII. MAGNETIC MEASUREMENTS. 44
VIII. THE SHIP'S MAGNETISM. Preliminary 52
IX. THE SHIP'S PERMANENT MAGNETISM 56
X. THE SHIP'S INDUCED MAGNETISM 60
XI. THE THEORY OF THE DEVIATIONS OF THE COMPASS 66
XII. TRANSFORMATION OF EQUATIONS (4)AND (5) 77
XIII. DEVIATIONS IN TERMS OF THE COMPASS COURSE 79
XIV. HEELING ERROR (Including Pitching Error) 86
XV. ANALYSIS OF DEVIATIONS 107
XVI. THEORY OF THE MECHANICAL CORRECTION OF THE
COMP ASS . 113
XVII. THE MECHANICAL CORRECTION OF THE COMPASS 145
ApPENDIX. Worked Examples 150
INDEX . 163
 DEFINITIONS OF TERMS, TABLE OF SYMBOLS
AND FORMULAE
Aclinic Line. The line on a chart through places where the value of the dip is
zero. Synonymous with Magnetic Equator.
Agonic Line. A line on a chart through places of no variation, that is to say,
where an undisturbed compass needle will point to geographical north.
Ampere per metre. The unit of magnetic field strength. One ampere per metre is the
strength of magnetic field inside a long solenoid wound with n turns of wire per
metre of its length, carrying a current I such that the product nI is one ampere-
turn per metre.
Coercive Force. The value of the reversed magnetic field required to destroy the
remanent or residual magnetism in magnetic material. It is a measure of the
coercivity or the tenacity with which magnetism is held in a substance.
Demagnetising Effect. This refers in a particular sense to the field produced by
the poles of a magnet in opposition to the field within it.
Equivalent length of a Magnet. The distance between the opposite poles. These
are not usually situated at the extremities of the material, and in the case of a
bar magnet are considered to be situated at approximately one-twelfth of its
length from each end. (See Magnetic Poles.)
Hysteresis. When iron or steel is magnetised by an external magnetic field which
is made to vary through a cycle of values, the magnetisation of the iron or
steel lags behind the field. This phenomenon is called hysteresis.
Induced Magnetism. A term used to describe magnetism induced in magnetic material
of low coercivity and remanence such as magnetically soft iron. The field
produced reduces to zero when the magnetising force is removed.
Intensity of Magnetisation. This is given by the magnetic moment divided by the
volume of the magnet and is therefore magnetic moment per unit volume.
Inverse Square Law. In all cases where an effect is radially and uniformly distributed
with respect to a point, the effect per unit area falls off inversely as the square
of the distance from the point.
Isoclinal. A line drawn on a chart through places having the same value of magnetic
dip.
Isogonal. A line drawn on a chart through places having the same value of magnetic
variation or magnetic declination.
Least Squares. The method of least squares is a method of close approximation
for obtaining the most probable value of a quantity from a set of physical
measurements. In the case of deviation analysis the principle of least squares
assumes that the best values of the coefficients are those which make the sum
of the squares of the errors a minimum.
Magnetic Coercivity. The measure of the tenacity with which magnetism is retained
by a substance.
Magnetic Declination. The scientific name for Magnetic Variation, q.v.
 Definition of Terms, etC.-continued
Definition of Terms, etc.-continued
Magnetic Dip. The angle measured in a vertical plane between the direction of the
earth's magnetic field at a place and the horizontal. Neutral Axis of a Magnet is an axis through it midway between the poles and
perpendicular to the axis through the poles. It is sometimes referred to as the
Magnetic Field. The space in which forces of attraction and repulsion due to magnetic Equatorial Axis.
effect may be detected.
Permanent Magnetism. The magnetism which is retained in magnetic material of
Magnetic Field Strength. The unit is the ampere per metre. The magnetic field
high remanence and coercivity for a long period of time. It can be destroyed
strength at any point in a magnetic field is the force that would be exerted on a
by heating, by violent physical vibration of the material, by the application
magnetic pole of strength one weber placed at that point.
of an opposing magnetic force of sufficient strength and to a lesser extent by the
Magnetic Flux. The total number of lines of force or induction crossing a given demagnetising effect of the field of the magnet itself.
surface area in a magnetic field is called the magnetic flux. The unit is the
weber. Ratio. The relation or proportion of one quantity to another.
Retentive Magnetism. See Sub-permanent Magnetism.
Magnetic Flux Density or Magnetic Induction. The unit is the tesla. Flux density is
defined as the number of lines of flux crossing an area of 1 sq. metre, the Sub-permanent Magnetism. A term used with reference to ship's magnetism. It refers
surface area being considered at right angles to the direction of the field. to the magnetism induced in magnetic material of medium remanence and
coercivity. Sub-permanent magnetism is retained for a shorter or longer period
Magnetic Inclination. An alternative name for magnetic dip.
of time depending on the remanence or retentive quality of the material.
Magnetic Moment. The magnetic moment of a magnet is the product of the pole
Weber. The unit of magnetic flux. It is the flux which, linking a circuit of one turn,
strengt\1 and the distance between its poles. It may also be defined as the
produces in it an electro-motive force of one volt as it is reduced to zero at a
couple required to maintain the suspended magnet at right angles to a magnetic
field of unit strength. uniform rate in one second.

Magnetic Permeability. The ratio of the magnetic induction in the material to the
strength of the magnetising field to which it is subjected, permeability of air or
vacuum being taken as 4n 10- 7
Magnetic Pole. The region of a magnet which exhibits magnetic properties from
which the greater part of the magnetic flux emerges or at which it enters. In
the case of a bar magnet the longer the bar in comparison with its thickness
the more nearly do the poles approach the ends of the magnet.
Magnetic Point Pole. A mathematical conception which may be considered as the
pole of an infinitely long and infinitely thin bar magnet.
Magnetic Pole Strength. The strength of a magnetic pole is equal to the flux
emerging from it. The unit is the weber.
Magnetic Remanence. The magnetic flux (magnetism) remaining in a magnetic
substance after the magnetising force has been removed.
Magnetic Screening. See Shielding.
Magnetic Shielding. The tendency of magnetic lines of induction to concentrate on
material of high permeability makes it possible to partly screen an area from
the effect of a magnetic field by interposing material of high permeability
between the source of the field and the area to be shielded.
Magnetic Susceptibility. The ratio of the intensity of magnetisation produced in a
substance to the magnetising force or intensity of field to which the material is
subjected.

Magnetic Variation. The angle between the vertical plane containing the direction
of the earth's field at any place and a vertical plane containing the geographic
north and south meridian.
Magnetometer. An instrument for making magnetic measurements.
Moment of Inertia. The moment of inertia of a body about any axis may be defined
as the sum of the products of all the elementary masses which make up the
whole body and the squares of the perpendicular distances of the elementary
masses from the given axis.
 CHAPTER I

INTRODUCTION

THIS book is written in an attempt to supply all the information

that is required for the efficient maintenance and correction of the
magnetic compass on board ships.
The methods employed in the mechanical correction of the
compass in the Merchant Navy, and for that matter in the Royal
Navy, are based on certain assumptions and approximations, the
accuracy of the results depending on how far these assumptions
are permissible in the actual conditions prevailing.
These methods rely on normal well-established laws relating to
magnetic fields and material, but it is not possible by empirical
means alone to ascertain the exact qualities and effects embodied
in all the magnetic fields in the ship, even though they may be
known to exist, nor to forecast all the individual effects, under
other conditions than those pertaining to a particular set of experi-
ments, in a suitable form so that a practical correction of the compass
may be made.
For instance, a thin soft iron bar, lying fore and aft alongside
and to starboard of the compass with its fore end abreast the
compass, would by experiment be shown to cause a maximum
westerly deviation on North and South magnetic and no deviation
on East and West magnetic indicating that the deviation from it
will vary as the cosine squared of the magnetic course when the
vessel is turned round in azimuth. Such a condition, if present,
is not conducive to mechanical correction. It can be shown mathe-
matically, however, that this condition may be split into two, one
involving a constant deviation on all headings of the ship which,
though not capable of correction, can easily be allowed for, and
the other involving a deviation varying as the cosine of twice the
course, which can be corrected by slewing the spheres.
Other examples could be quoted, and it is for reasons such as
this, together with the fact that approximations are involved, that
it is essential to have some conception of the mathematical theory
underlying the mechanical correction of the compass if one is to gain
the confidence that comes from an understanding of the subject
rather than from a superficial knowledge of it.
2 INTRODUCTION
The author has, as far as is possible, borne in mind the varying
degree of knowledge which may be required by those who for one
reason or another are interested in the subject. It has also been
written in as simple language as possible omitting none but the
most obvious of details even to the extent of some repetition.
In all cases it is suggested that the book should, initially, be
read right through as a paper of technical interest, omitting chapter
XII and the mathematics at the end of chapter XIV. By so doing
the reader will obtain a mental picture of the whole field covered
together with the reasons for the final results and at the same time
discover which sections require his particular attention should the
complete detail be more than he requires for his more immediate
needs.
It is submitted that those teaching the subject will require to
know all that the book contains. This also applies to compass
adjusters who wish to have a thorough grasp of the subject, bearing
in mind that the safety of the ship depends to some extent on their
proficiency. It should also be of considerable assistance to those
studying for the Extra Master's Certificate, as there is little con-
tained in the book which could not be read into the syllabus,
remembering that although the rods b, d,f, and h are for some reason
specifically omitted, the candidate is still expected to have a full
understanding of the various coefficients which are, after all, only
abbreviations for combinations covering all the rods together with
the forces P, Q and R.
For the Certificate of Competency as Master a great deal will
depend on the time available for study. It would not be possible
to cover the whole work in say two months study time ashore,
bearing in mind the other subjects to be taken. The minimum
required for the certificate is covered by Chapters IV, VI, VIII,
IX, X, the results obtained in Chapters XI and XIII, pages 98 to 102
of Chapter XIV on heeling error, the rules at the end of Chapter XV,
Chapter XVI except the section on errors of the deflector, and
Chapter XVII.
It is suggested that by short periods of fairly regular study at
sea the whole of the subject could be easily mastered in the course
of a few voyages.
Apart from the minimum knowledge required to obtain a Certifi-
cate of Competency, the increasing number of navigational aids
being fitted in ships in the vicinity of the compass, having magnetic
fields of their own which may affect the compass and which may
not be symmetrical with reference to its position, make it advisable
INTRODUCTION 3
that this subject should be more easily understood by the Masters
and Senior Officers of ships.
In the author's opinion the most tiresome and complicated part
of the study is that dealing with heeling error and in particular
the effect of the various rods when the ship is heeled. It has been
dealt with in full detail in the chapter on heeling error, but if the
reader provisionally accepts equation lla on page 97 he can proceed
from that point leaving the earlier sections to be studied later. This
also applies to the mathematics at the end of the chapter. Chapter
XII may also be reserved for separate study.
Chapter V describes the fundamental facts of magnetism needed
in connection with the rest of the work. A fuller exposition, if
required, may be found in any text-book on elementary magnetism
and electricity.
Chapter VII on magnetic measurements has been included as
it is thought that the book would not be complete without some
reference to the method of finding the various magnetic values
referred to.
Chapters II and III may be considered superfluous by the more
mathematically minded, but the trigonometrical relationships at
the end of Chapter III will most certainly be required. A know-
ledge of the signs of the trigonometrical ratios in the consecutive
quadrants is also very important as by their use, together with
the positive sign for easterly and the negative sign for westerly
deviation respectively, all the problems involving deviation may
be worked without any "rule of thumb" or aids to memory once
a knowledge of the theory has been gained.
Fundamentally, most problems are as simple as the following:
"If six herrings cost eighteen pence, how much will eight cost?"
To solve this problem one must first know that the cost varies
directly as the number of herrings. We first find the cost of one
by dividing by six and then multiply by eight, the answer of course
being twenty-four pence.
In deviation problems the deviation represents the cost and
the factors on which it depends are the herrings. For instance,
it will be found in a later chapter that a force P causes a deviation
which varies inversely as H and directly as the sine of the course,
so that if P causes 6 westerly deviation on N 30 E at a place
where H = 16, we can find out what deviation it will cause on
say, N 45 W at a place where H = 32. Thus the cost of 030
coupled with 16 is _6. To find the cost of one of each of these
coupled together we must divide by the sine of 030 and also by
 CHAPTER IV

MAGNETISM
General

THE Molecular Theory of Magnetism is universally accepted, though

it has been modified from time to time from its original form.
This theory states that all magnetic substances consist of magnetic
molecules each being a minute magnet. These magnetic molecules
are not necessarily chemical molecules, and it is not within the
scope of this book to discuss their precise nature, which is, in the
light oflater investigations, atomic rather than molecular. When the
substance is unmagnetised these minute magnets are not arranged
in any particular direction, but are orientated indiscriminately in
all directions. It can in fact be proved that they prefer this arrange-
ment rather than to be lined up in a particular direction. Once
these minute magnets are aligned, however, the mutual attraction
of their poles tends to hold them in position after the removal
of the external force used to align them. This alignment can be
destroyed by physical vibration such as hammering, or by heating.
The ease with which it can be destroyed depends on whether the
substance is magnetically hard or magnetically soft.
In the case of ferrous material, the substance in which we are
most interested, the terms hard iron and soft iron refer specifically
to this particular property.
In pure soft iron the molecules are entirely free and, unless under
the influence of some external magnetic field, the iron will be non-
magnetic or un-magnetised.
This is not so in the case of hard iron. The molecules are not
free to move nor are they easily moved, but once lined up in a
particular direction by an external force they tend to remain in
that direction indefinitely, and the iron is said to be permanently
magnetised. It will suffice to mention here that these so-called
permanen t magnets are, apart from the effects of vibra tion or heating,
not truly permanent, but tend gradually to lose their magnetism
during the course of time. Magnets made of magnetically hard
substances are normally referred to as permanent magnets, and for
the purposes of this book the effects of all iron of this nature will
be covered by the term permanent magnetism.
19
20 MAGNETIC COMPASS DEVIATION AND CORRECTION MAGNETISM 21
of the ship's head will be kept sufficiently small for all practical
What, then, of the material used in shipbuilding? The major
purposes.
portion is of mild steel. This material cannot be considered as
There are, however, circumstances under which a ship may have
either magnetically hard or magnetically soft, it has in fact both
her magnetic field considerably altered both in direction and
these properties as well as every intermediate type that can be
amount, and which may render earlier or more frequent adjustment
considered lying between the two. Some groups of the molecules
necessary. This may be if the ship has been struck by lightning
act entirely as in the case of those of soft iron, while other groups
(see note at end of chapter), or if her cargo has been loaded or dis-
are progressively less easily disturbed, till we find that some behave
charged by means of electro-magnets, or if the ship has suffered
exactly as the molecules of hard iron. This complicated state of
severe damage, as in collision, or has undergone material structural
affairs would at first appear to make the investigation of its effects
alteration. The carriage of relatively large quantities of magnetic
on the compass very involved, but fortunately this is not the case.
material as cargo, such as steel rails or some types of iron ore may
It is possible to separate the soft iron effects and the hard iron
also cause marked changes in the deviations of the compass. When
,effects,and as far as the investigations are concerned, the effects
steering one course for a considerable time the ship may also acquire
appear as if there were in fact soft iron material and hard iron
sub-permanent magnetism, of which the poles will be in the magnetic
material in the ship as separate entities, together with material the
meridian relative to her head.
effects of which lie somewhere between the two. These latter effects
The ship's officer must therefore realise that although the ship's
will subsequently be referred to as being due to sub-permanent
compasses may be regularly adjusted by a qualified compass
magnetism. In some text-books this is referred to as retentive
adjuster, he must always be on the alert and be able to anticipate
magnetism.
changes that may occur in the deviations, due to limitations in the
The term effects continually recurs in the text of this chapter,
method of adjustment or due to circumstances over which the
and it must be emphasised that though we know that the material
compass adjuster has no control.
of which the ship is built is chiefly mild steel, and that though the
Note. In January 1949 while on station in the North Atlantic
effects obtained indicate that there is apparently both hard and
the S.S. Weather Recorder was struck by lightning during a severe
soft iron present, the effects are real and continuously there and
storm. Previous to being struck the maximum deviation of the
are the cause of the deviation or deflection of the compass needle
standard compass on any heading was 31, after the incident the
from the direction of Magnetic North. It is these effects which we
deviations were found to range from a minimum of 4 to a maximum
measure in terms of the deviation caused and for which we com-
of 31 as shown in column 1 of the analysis of the ship's deviations
pensate, using permanent magnets suitably disposed to compensate
given on page IlIa to which the reader is referred. Subsequent
for the hard iron effects and soft iron suitably disposed to compensate
observations indicated that the deviations, in this particular case,
for the soft iron effects.
took about six months to settle down so that during this time
It is the case, unfortunately, that at any given time of investigation
frequent adjustment of the correctors was required. The above
the sub-permanent effects appear to be permanent and merely
information was obtained from, and permission to publish kindly
modify temporarily the permanent effect for which compensation
given by, the Director of the Meteorological Office of the Air
is made. It is chiefly for this reason that a ship requires to
Ministry, Harrow, Middlesex.
be swung periodically to have the positions or the number of
the correctors used in previous compensations suitably adjusted to
allow for the change in this sub-permanent part of her magnetic
field.
As the change in the sub-permanent magnetic field is normally
relatively slow, so that the increase in the deviations caused by
this change is not rapid, it does not involve a very frequent adjust-
ment of the correctors. Usually if the ship is swung for adjustment
at intervals of about twelve months the deviation on any direction

CHAPTER V
FUNDAMENTAL FACTS OF MAGNETISM
Magnetic Poles. If a bar magnet is dipped into iron filings it will
be found that the filings adhere most freely at and near the ends
of the bar. These places are called the poles of the magnet.
If the magnet be suspended freely it will be found that one of
the poles will always point to the Magnetic North providing no
other magnetic field is present. This pole is called the north seeking
pole or north pole of the magnet. The other pole is called the
south seeking pole or the south pole of the magnet. The north
seeking end of the magnet is often painted red and marked with
a letter N, the south seeking end is then painted blue and marked
with a letter S. When paint is not used and when only one end is
marked, it is invariably the north end. A line drawn through the
magnet in the direction of its internal field, i.e., joining its poles, is
called the magnetic axis of the magnet, and a line at right angles to
this direction midway between the poles is called the neutral axis
of the magnet.
Magnetic poles exert a force upon one another. Like poles repel,
and unlike poles attract one another. The force between two poles
is dependent upon their distance apart.
The strength of the north and south poles of a magnet is equal.
If a bar magnet is placed upon a float in a bowl of water so that it
is free to move in any direction, it will be found that the magnet
will rotate with its float until the north seeking end points north
and, of course, the south seeking end points south, but it will not
move either toward the north or south. The attraction of the north
end toward the north is therefore exactly equal to the attraction
of the south end toward the south.
Magnetic Field. In general terms the magnetic field of a magnet
may be defined as a region round it throughout which its influence
may be detected. The term, however, is usually reckoned to have a
more definite and restricted meaning. If a single magnetic pole
could be placed at any point near a magnet it would experience a
force in a definite direction, and if free to move it would travel
in this direction. This is called the direction of the magnetic field
at that point.
22
FUNDAMENTAL FACTS OF MAGNETISM 23
If a very short, freely suspended, compass needle be brought near
to the end of a bar magnet, it will be found that one end of the
needle will point directly toward the pole of the bar magnet, and it
can be shown that the magnetic field radiates in all directions from
the pole away from the length of the magnet.
If the needle be brought vertically over one pole of the bar
magnet lying on a table, the needle will take up a vertical position,
and if carried slowly along the length of the magnet toward the
other pole it will gradually tilt out of the vertical until at mid length
it will be seen to be horizontal. As the other end of the bar magnet
is approached the opposite end of the compass needle will start
to dip below the horizontal and continue to do so increasingly
until it reaches the other pole, when it will be seen to stand vertical
again with its opposite end downward. The compass needle may
be said to have followed the direction of the lines of force of the
magnet between its poles.
Lines of force are said to arise on a north pole of a magnet and
end on a south pole. No two lines of force can meet or cross one
another.
It will be seen later that this conception of lines of force may
be used as a quantitative representation of the strength of the
magnetic field at any particular point in it. It is a useful idea in
so far as it helps one to visualise the field in a definite manner.
Magnetic Pole Strength. The pole strength of a magnet is con-
veniently defined as the number of unit poles acting together which
are required to make up an equal pole strength.
The Inverse Square Law. It can be proved that in all cases where
an effect is radially and uniformly distributed with respect to a
point, the effect per unit area falls off inversely as the square of the
distance from the point.
The effect on a poi!1t magnetic pole due to another point magnetic
pole follows the same law. This can be proved experimentally by
means of a magnetometer.
We can say therefore: 'The force between two point poles varies
directly as their pole strengths and inversely as the square of their
distance apart."
Magnetic Intensity or Magnetic Field Strength. It has been shown
that a magnetic field has a definite direction at every point in it.
In order to define the force at any point in the field it is necessary
to think of its effect on an imaginary unit pole placed at that point.

CHAPTER VI
THE EARTH'S MAGNETISM
THAT the earth has a magnetic field is obvious from the fact that
a freely suspended magnet will come to rest in a direction approxi-
mately North and South. In other words, the magnet will settle
in a direction of the earth's field at the place at which the magnet
is being used.
At most places on the earth's surface this direction will not be
horizontal nor will it liein the geographical meridian. In the British
Isles the north-seeking end of a freely suspended magnet will dip
below the horizontal and be inclined at an angle to the westward
of True North.
In general, if the magnet were carried northward over the earth's
surface following the surface direction of its north-seeking end, we
should find that the magnet would dip at an ever-increasing angle
from the horizontal until, eventually, we should arrive at an area
where the magnet would take up a vertical position and have no
directional value whatever. On the other hand, if we proceeded
southward following the south-seeking end, we should find that
the angle from the horizontal would gradually decrease until, when
somewhere in the vicinity of the geographical equator, the magnet
would lie in an exactly horizontal position. Continuing southward
the south-seeking end would dip below the horizontal until
eventually we should reach another area where the magnet would
assume a vertical position, this time with the south-seeking end
downward.
This imaginary experiment would, therefore, give similar results
to the experiment described on page 23, when the effect of passing
a small compass needle over the length of a bar magnet was noted.
In fact the area on the earth's surface where a freely suspended
magnet would set vertically with its north-seeking end downward
is called the North Magnetic Pole, and that area where such a
magnet would set vertically with its south-seeking end downward
is called the South Magnetic Pole.
It should be noted that the north magnetic pole of the earth
is of the same affinity as the south pole of the magnet, and would
conventionally be coloured blue. The south magnetic pole of the
36
THE EARTH'S MAGNETISM 37
earth is of the same affinity as the north pole of the magnet and
would conventionally be coloured red.
It would thus appear that the earth's magnetic field is similar
to that of a bar magnet. As a first approximation this is substantially
correct. The general magnetic field of the earth is similar to that
which could be expected at the surface if a short but strongly
magnetised bar magnet were located at the centre. This partly
explains the fact that the magnetic poles are relatively large areas,
due to the spreading out of the lines of force from the magnet, it
also gives a reason for the direction of the field being horizontal in
the vicinity of the equator. It is most improbable, however, that
there is such a magnet at the centre of the earth, and in actual fact
the cause of the field is still being investigated by many learned
scientists. No theory put forward up to the present has found
general acceptance.
As far as we are concerned, the idea of the magnet at the centre
of the earth is useful as it helps us to visualise the general form of
the magnetic field as it is known to be despite the many imperfec-
tions which will become apparent as we proceed.
The area termed the North Magnetic Pole is situated in approxi-
mately 71 N., 96 W. and the South Magnetic Pole in 73 S.,
156 E. These positions are very approximate, but one fact emerges,
namely, that the south pole is not diametrically opposite to the
north pole.
A line joining all positions on the earth's surface where the direc-
tion of the magnetic field is horizontal is called the Magnetic Equator
or Line of No Dip. It is not a perfect circle but a wavy line
which crosses the geographical equator in two points, from north
to south in about 35 W. and from south to north in about 170 W.,
reaching its most northerly position in central Africa and most
southerly position in Brazil.
Theoretically the maximum strength of field should be at the
poles, actually the field strength in certain other areas in both high
north and south latitudes is found to exceed that at the magnetic
poles. These areas are called Magnetic Foci.
In order to determine the direction and force of the earth's
magnetism at any place we require to know three of four magnetic
elements. The four elements are Variation, Dip, Horizontal Force
and Vertical Force.
Magnetic Variation or Declination at a place on the earth's surface
is the horizontal angle contained between the true meridian and the
 38 MAGNETIC COMPASS DEVIATION AND CORRECTION
direction of the lines of force of the earth's magnetic field at the
place. It should be noted that this latter direction is not necessarily
in the exact direction of the magnetic pole from the place; this is
due to slight distortions in the earth's field of a semi-local character.
This fact is verified if one examines a Variation chart.
Linesjoining places on a chart having the same value of Variation
are called Isogonal Lines. Lines drawn through places where the
Variation is zero are called Agonic Lines.
When the direction of the earth's field inclines to the left of True
North, that is when a compass needle, under the influence of the
field at the place, points to the West of True North, the Variation
is said to be Westerly. When the inclination is to the right of True
North the Variation is said to be Easterly.
Variation at all places on the earth is undergoing a long period
change. One theory suggests that the magnetic poles revolve round
the geographical poles once in about 960 years, but the rate of
change from earlier observations does not appear to be constant.
Variation also goes through annual and daily alterations in its
value, these are small, however, and do not affect the navigator.
Variation charts, showing Isogonal lines covering the world and
indicating the mean annual change, are published by the British
Admiralty at intervals of about five years.
In order to measure the variation it is necessary to ensure that
no local magnetic field is present. If conducting such a measurement
at sea, it would be essential that the whole ship were built of non-
magnetic material. A special magnetic needle is required and, for
accurate observations to minutes of are, an instrument such as the
Kew Magnetometer is used. The bearing of a suitable object is
obtained with the instrument, the needle of which is then turned
over and the observation repeated. The mean of the bearings is
then compared with the known or calculated true bearing of the
object, and the Variation found. The reason for turning the needle
over is that its magnetic axis may not be parallel with the geometric
axis. If the line joining the poles of the needle is not parallel with
the geometric or visual axis, the needle will point to one side of
the magnetic meridian at the place, but when turned over will point
exactly at the same amount on the other side of the magnetic
meridian. The mean of the two directions will therefore be the
correct direction of the magnetic meridian at the place. For ordinary
purposes of navigation, if the standard compass is well placed
magnetically, the variation may be ascertained by swinging the ship
slowly in azimuth and observing the compass bearing of a celestial
THE EARTH'S MAGNETISM 39
body, suitably placed, on eight or more equidistant points. The
mean of the differences between the compass bearings and the
corresponding true bearings will be the approximate variation.
Magnetic Dip or Inclination. The vertical angle contained between
the horizontal and the direction of the earth's magnetic field at
any given place is called the Angle of Dip. Dip is conventionally
considered positive when the north end of a freely suspended
magnetised needle dips below the horizontal, and negative when
the south end dips below the horizontal. Hence all angles of dip
north of the magnetic equator will be positive and all angles of
dip south of the magnetic equator will be negative.
Lines drawn on a chart joining all places for which dip has the
same value are called Isoclinals. The line joining all places where
no dip occurs is called the Aclinic Line; it is also referred to as
the Magnetic Equator, and has already been described.
The value of the angle of dip at all places undergoes similar
changes to those described for variation. The cause of the changes
is not yet entirely understood. Charts of the world, showing lines of
equal dip and indicating the mean annual change, are published
by the British Admiralty in the same way as Variation charts.
Dip is usually measured by means of an instrument called a
Dip Circle. This consists of a vertical circle and scale at the centre
of which the dip needle is suspended on a fine steel axle resting on
two agate knife edges. The stand contains a horizontal azimuth
circle to facilitate placing the magnet in the plane of the magnetic
meridian. A spirit level and levelling screws are fitted to the stand.
Low-powered microscopes with cross wires and verniers are fitted
to the framework for greater accuracy of measurement.
Once the instrument has been levelled and its plane placed in
the magnetic meridian there are four potential sources of error to
be allowed for, involving the making of sixteen observations:
(1) The axis of rotation of the magnet may not be at the centre
of the scale. Both ends of the pointer are therefore read and the
mean of the readings taken.
(2) The zero line of the circle may not be horizontal. By rotating
the dip circle through 180 this error is reversed, but the mean
0
of the upper and lower readings must again be taken and a mean
of the means made.
(3) The magnetic axis of the magnet may not coincide with its
geometric axis. The magnet must therefore be reversed on its
bearings and the four readings obtained by (1) and (2) repeated.
 THE EARTH'S MAGNETISM 41
if required or simply eliminated by the simple algebraic process
of combining the simultaneous equations formed from the results
of the two experiments. The full procedure is described in the next
chapter.
Both Hand Z may now be measured by more modern electrical
instruments called coil galvanometers. These instruments can be
specially designed to measure H or to measure. Z, but are not
entirely satisfactory, as they are difficult to calibrate accurately,
though it is probably only a matter of time before they entirely
replace the magnetometer for this purpose.
The strength of the earth's magnetic field is subject to periodic
changes covering long and short periods of time, but little is known
of the cause or precise value of the long-period change. It is known
that when sunspot activity is pronounced short-period fluctuations
are more marked, and that on some days, called magnetically quiet
days, the elements undergo smooth regular variations, while on
others, called magnetically active days, they are more or less dis-
turbed. There appear also to be fluctuations due to daylight hours
and due to dark hours as well as to a lunar cycle. These minor
fluctuations in the values of the elements are small and do not
affect the navigator. As in the case of Variation and Dip, charts
of the world are published showing the values of Hand Z for all
places on the earth's surface together with the mean annual change.
These values are now given in micro-teslas, units of magnetic
flux density. A micro-tesla is one millionth part of a tesla. To convert
flux density in micro-teslas to magnetic field strength in ampere
metres the following conversion factor should be used:
 THE EARTH'S MAGNETISM 43

its normal direction so that, for instance, a large Easterly Variation

may change rapidly to a large Westerly Variation as the spot is
approached. The general tendency of these disturbed areas is to
attract the north-seeking end of the compass needle in North
Magnetic Latitudes and to repel the north-seeking end of the needle
in South Magnetic Latitudes. It is for this reason that these areas
of local magnetic disturbance are often somewhat loosely referred
to as areas of local magnetic attraction.
The value of the Dip and of the intensity of the magnetic field
is also considerably affected in these localities, the former may be
as much as 30 greater than in the immediate vicinity clear of the
local field, while the horizontal field may be so reduced as to be
ineffective at the compass position.
 THE SHIP'S MAGNETISM 53

Magnetic North, or more accurately, away from the direction of the

earth's field at the place, when the compass is placed on board the
ship.
A line drawn in the direction of the earth's magnetic field at a
given place is usually termed the magnetic meridian at the place,
and is analogous to the term geographical meridian.
The angle through which the compass needle on board ship is
deflected from the magnetic meridian due to the ship's magnetic
field is called the Deviation of the Compass. If the north end is
deflected to the eastward or clockwise, the deviation is said to be
Easterly, and if deflected to the left or anticlockwise, the deviation
is said to be Westerly. Easterly deviation is also considered positive
and Westerly deviation negative.

Mariner's Compass. The term compass needle is meant to cover

the various systems of magnetic needles found in compasses used
on board ships.
This type of compass, referred to as the Mariner's Compass, is
used to find the direction of the ship's head relative to the magnetic
meridian and to ascertain the bearings of terrestrial and celestial
objects from the ship. The compass needle is suspended so as to
have freedom of movement in the horizontal plane only, it is there-
fore affected by the horizontal component of the earth's field alone.
In order to prevent the roll of the ship from setting up oscillations
of the compass card, the disposition of the mass of the card in the
various directions from the pivot is taken into account in its manu-
facture.
It has also been determined mathematically that the proper system
of needles is that in which the mean of the angles subtended at the
centre of the compass by symmetrical pairs of like poles is 60. This
arrangement also happens to be the correct arrangement magneti-
cally, as it gives the equivalent of a short magnet placed at the centre
and helps considerably to ensure that a satisfactory correction can
be made.
The bowl containing the card is placed in the compass binnacle,
which is designed to allow the bowl free movement in a gimbal
system, so arranged that the compass remains horizontal for all
movements of the ship without appreciably displacing the lubber
line from the fore and aft line of the ship.
Binnacles are constructed of non-magnetic material, usually wood,
and are of such a shape, size and height that bearings of terrestrial
or celestial objects may be taken most conveniently. They are fitted
54 MAGNETIC COMPASS DEVIATION AND CORRECTION THE SHIP'S MAGNETISM 55
to receive the various correctors used to counteract the ship's needle, and the latter decreases the directive force. The increase of
disturbing magnetic fields. directive force is in some respects desirable but the inevitable
The correctors consist of a Flinders Bar, a bar of soft iron placed corresponding decrease is not. Further, in order to ensure reliability
vertically on the outside of the binnacle; spheres, soft-iron solid or of the compass, it is necessary that it should remain in a uniform
hollow globes, are also fitted on the outside of the binnacle; a field in the direction of the magnetic meridian as the ship rotates
Heeling Error Bucket capable of being moved vertically up or down round it.
in the centre of the binnacle and which is fitted to hold permanent The principle underlying the correction of the compass can there-
magnets in a vertical position. The binnacle is also fitted to accom- fore be stated as follows: If the directive force at the compass position
modate fore and aft and athwartship horizontal permanent magnets. be equalised on all points of the ship's head in azimuth there will be no
deviation on any point, and any alteration of the ship's head in azimuth
Magnetic Field at the Compass Position. The direction of the will be correctly recorded by the compass.
ship's magnetic field at the compass position will not necessarily be
that due to the direction of the ship's head when building, but will
be the resultant of several fields of which the general tendency as
indicated in Figs. 24 and 24Amayor may not preponderate. A part
of the ship mass-produced may acquire a magnetic field, the direction
of which, when such part is placed in position, may differ consider-
ably from the direction of the field of the rest of the material in the
vicinity. Local fields may also be set up in the vicinity of welded
parts. The magnetic field due to welding when Direct Current is
used is known to lie at right angles to a line joining the electrode
and the earth and to follow the same rule as for a current flowing in
a wire. One of the most important influences on the ship's field at
the compass position is that due to the disposition of Vertical Hard
Iron effects in that locality. This is discussed more fully in the next
chapter.
The resultant magnetic field of the ship at the position of the
compass changes its direction with respect to the compass needle,
as well as its value, with every change of direction of the ship's head,
consequently the deviation will alter with each change of the
direction of the ship's head. "This means that any alteration of the
direction of the ship's head will not be correctly recorded by an
uncorrected compass, a circumstance which, if the deviation were
large, might have disastrous consequences especially in crowded
waters. The fact that when bearings are taken by the compass the
deviation for the particular direction of the ship's head must be
allowed for, and also that magnetic courses have to be corrected for
deviation to obtain the compass courses to steer and vice versa, is
a minor consideration.
The ship's field will, on certain ship's heads, be acting more or
less with the earth's field, and on others be acting against it. The
former condition increases the directive force acting on the compass
 CHAPTER IX
THE SHIP'S PERMANENT MAGNETISM
IN order to compensate for the ship's disturbing magnetic fields the
method adopted is to divide the effects into components and then
deal with each component separately. The components chosen are
those related to the fore and aft, athwartship and vertical directions
in the ship through the centre of the compass needle system.
These components are termed the Fore and Aft component, the
Athwartship component and the Vertical component of the ship's
field respectively, and when considered conjointly are always written
and thought of in that particular order. Any force under consider-
ation which is acting at an angle to any of these direCtions is first
divided into two or three of these components. Reference to this
method has already been made in Chapter I.
The division of the earth's total force into horizontal and vertical
components has also been discussed.
The ship's permanent magnetism may be defined as that part of
the ship's magnetism which would remain unaltered if we imagine
the ship to be entirely removed from any external field, and in
particular the earth's field.
As previously stated, the sub-permanent magnetism of the ship
together with that of a lasting nature are treated as permanent for
the purposes of correction.
That part of the ship's permanent magnetism acquired due to
the influence of the earth's field, if horizontal in direction, may be
considered due to the earth's horizontal component H, and that
which is vertical may be considered due to the earth's vertical
component Z.
The effect due to Z in the northern hemisphere is to cause a blue
pole at the upper end of vertical structures and a red pole at the
lower end. This reasoning applies, for instance, while the ship is
being built and is irrespective of the direction of the ship's head on
the stocks or the position of the structure relative to the compass
position.
Consider the effect of such a structure on deviation (see Fig. 25).
The permanent blue pole of the structure will cause an attraction
of the north end of the compass needle forward and downward in
56
having exactly the same effect. As the compass card is constrained
to the horizontal, the vertical component will have no effect on
deviation when the ship is upright. If the structure i~ ific the centre
line of the ship, the horizontal component will cause no deviation
when the ship is heading north or south by compass, but will cause
an easterly deviation on easterly courses and a westerly deviation
on westerly courses.
It will thus be realised that the effect of a permanently magnetised
vertical structure on deviation is exactly the same as if it were
horizontal with its same pole (in this case the blue pole) nearest
the compass and its length in the line joining the centre of the
compass to the near pole.
The effect of all forces under consideration is measured in terms
of the deviation which they cause, thus it is obvious that the amount
of vertical permanent magnetism in any direction from the compass
position is indeterminate, as the deviation it causes might just as
effectivelybe caused by permanently magnetised horizontal material
of the same polarity in the same vicinity.
It is for this reason in particular that the general tendency as to
the direction of the ship's magnetic field, discussed in the previous
chapter and illustrated in Figs. 24 and 24A, may be considerably in
error.
Next consider a section of permanently magnetised horizontal
material not in the horizontal plane through the compass as in
Fig. 26 (overleaf).
The near blue pole will cause an attraction of the north end of the
compass needle forward and downward in direction. Two com-
ponent forces, one horizontal and one vertical (shown dotted in
Fig. 26) could be substituted having exactly the same effect. As
before, the vertical component will have no effect on deviation when
 THE SHIP'S PERMANENT MAGNETISM 59
keel, and negative (- R) when this attraction is upward or toward
, the mast-head.
The method of determining and using the forces P, Q and R
together with the soft-iron effects in the investigation of deviations
of the compass is the subject of a later chapter.

the ship is upright. If the section is considered fore and aft in the
centre line of the ship, the horizontal component will have the same
effect on deviation as the fore and aft horizontal component of the
vertical structure discussed in the previous example.
The effect, therefore, of the ship's permanent magnetism on
deviation is such that, while the ship is upright, the deviation
appears to be caused entirely by horizontal permanently magnetised
material lying in the horizontal plane through the compass. The
vertical effect appears to be due to permanently magnetised vertical
material immediately below the compass causing heeling error when
the ship is inclined from the upright. Strictly speaking, one should
say above or below the compass, through a red pole above the
compass can be considered as a blue pole below the compass, and
vice versa.
The horizontal effects of the permanent magnetism, if not in a
fore and aft or in an athwartship direction, must be divided into
those two components. The fore and aft component of the horizontal
effect is referred to as the Fore and Aft Component of the ship's
permanent magnetism and is called force P.
Force P is considered positive ( + P) when it causes an attraction
of the north end of the compass needle toward the bow, and negative
(- P) when this attraction is toward the stern.
The athwartship component of the horizontal effect is referred to
as the Athwartship Component of the ship's permanent magnetism
and is called force Q.
Force Q is considered positive ( + Q) when it causes an attraction
of the north end of the compass needle toward the starboard side,
and negative (- Q) when this attraction is to port.
The vertical effect above or blow the compass is referred to as
the Vertical Component of the ship's permanent magnetism and is
called force R.
Force R is considered positive (+ R) when it causes an attraction
of the north end of the compass needle downward or toward the
 CHAPTER X

THE SHIP'S INDUCED MAGNETISM

IT has already been stated that there is an effect present at the

compass position similar to that which could be caused by magneti-
cally soft iron. The effect is dealt with in a similar manner to that
described for the effect of the permanent magnetism in the ship,
that is to say, it is divided into a fore and aft, athwart ship and
vertical component.
The effect is due, however, to induction by the earth's field in
soft iron, and the direction of the induction in the material will vary
with every change in the direction of the material with reference
to the earth's field. In other words, the direction of the induction
will change with every change in the direction of the ship's head.
This necessitates giving each of the components a fore and aft,
athwartship and vertical component to itself, making nine compon-
ents in all.
It might be argued in this case that the further division into
components could go on indefinitely, but this is precluded by de-
fining precisely the components in such a way that all the effects
of induction in material of a soft-iron nature are covered by nine
such components, no less but no more.
These nine precisely defined components are purely a mathe-
matical device, but it must be stressed that they do cover all the
practical effects and are capable of being easily understood.
Each component is considered to be a soft-iron rod without
thickness, so that it cannot be magnetised except through its length,
and of infinite length, so that only the near end or pole is effective,
i.e., able to act on the compass.
Certain of the rods are required to be imagined as passing through
the compass position. In this case it will be found that either end
may be considered as the near or effective end.
The nine rods are known by the letters a, b, c, d, e, f, g, hand k.
They should be thought of in groups of three in the alphabetical definition, a must be a fore and aft rod, b must be an athwartship
order given above. rod, and c a vertical rod.
The first three, a, band c, are the fore and aft, athwart ship and The second or middle group of three rods, d, e and f, are the
vertical components respectively of the fore and aft horizontal fore and aft, athwartship and vertical components respectively of the
component of the ship's induced magnetic field. Consequently each athwartship horizontal component of the ship's induced magnetic
60
 62 MAGNETIC COMPASS DEVIATION AND CORRECTION

field. Consequently, each of the rods must have its near or effective
end in the athwartship plane through the compass and, by virtue
of the definition, d must be a fore and aft rod, e an athwartship
rod, and f a vertical rod.
The last three rods, g, hand k, are the fore and aft, athwartship
and vertical components respectively of the vertical component of
the ship's induced magnetic field. Consequently, each of the rods
must have its near or effective end vertically above or below the
compass position and, by virtue of the definition, g must be a fore
and aft rod, h must be an athwartship rod, and k a vertical rod.
The foregoing Table of diagrams indicates the positions of the
various rods with reference to the compass position. Although
alternative positions are given and a positive or negative sign is
attached to each, it must be emphasised that the rods represent
effects and as such there need be considered but one of each of
the rods in the ship. Which of the alternative positions is used is
immaterial providing the sign is right. The method of determining
the sign of a rod is simple, as will be seen from the explanation
which follows below.
There is no need to memorise the Table, remember that the rods
are components, that the order is always "fore and aft, athwart ships
and vertical" in groups of three alphabetically arranged with i and
j omitted. Incidentally, it is much easier to keep all the nine rods
in mind than to single out only those which appear normally at
the position of a well-placed compass.
To Determine the Algebraic Signs of the Various Rods. It will be
remembered that in the case of a ship's permanent magnetism it
was shown that an attraction forward or to the bow was positive,
an attraction to starboard was positive and an attraction downward
or toward the keel was positive, while an attraction aft, to port or
upward was considered negative.
In the case of the rods it is convenient to remember simply that
"forward is plus, to starboard is plus and downward is plus", while
aft, to port or upward are negative.
Consider the direction of the near end of each rod from the
compass position and apply the appropriate sign (+ or -), then
consider the direction of the far end of each rod to its near-end
position and apply the appropriate sign (+ or -). The "product"
of the signs thus obtained will give the correct sign for each
particular rod.
Take, for example, the two positions of the +a rod given in the
Table and reproduced in Fig. 27.
 CHAPTER XVI

THEORY OF THE MECHANICAL CORRECTION

OF THE COMPASS

THE mechanical correction of that part of the deviations caused

by the ship's permanent magnetic field is effected by means of
permanent magnets, and is in practice straightforward, but in the
case of the soft-iron correctors, complications arise due to the fact
that the induced magnetism in the correctors may not be wholly
due to the earth's field. It may in part be due to the permanent
correcting magnets or due to the compass needle system itself.
Flinders Bar. The Flinders bar is used to correct that part of
the semicircular deviation caused by the excess of vertical soft
iron forward or abaft the compass, but symmetrically disposed to
starboard and port, i.e., c rod effect.
The bar is placed, with its long axis vertical, in a container
fixed usually on the fore side of the binnacle to correct a negative c.
The upper end of the bar is placed so that one-twelfth of its length
is above the horizontal plane through the compass-needle system.
The amount of correction is varied by using longer or shorter
lengths of bar. The effect of a shorter bar is less, partly due to
the increased proximity of the lower end and partly due to the
greater demagnetising factor of a shorter bar. (See page 26.)
As the breadth of the bar is not negligible, it has the effect of
+ a and - e rods causing a small coefficient D which increases the
normal + D of the ship for which the spheres are used to compensate.
Induction in the bar by the horizontal permanent correcting
magnets merely augments the corrections for which these magnets
are used. Provided, therefore, the bar is in position when the
magnets are placed, the compensation will be satisfactory from
this point of view.
Nevertheless induction due to the vertical permanent magnets,
used in the heeling-error correction, cannot be considered constant,
as the raising or lowering of these magnets to adjust the heeling-
error correction will alter the amount of induction in the bar and
upset the corrections made by the fore and aft magnets. This effect
is by no means negligible. A bar 24 inch (61 cm) in length has been
known to cause 9 deviation due to induction from a full "bucket"
113
 130 MAGNETIC COMPASS DEVIATION AND CORRECTION

-RN, RE, Rs, Rw, RNE, etc.-should first be multiplied by the cosine
of the appropriate deviation for the particular direction of the ship's
head involved.
Summary
(1) In the case of the Thompson deflector, it is essential that the
compass shall be deflected through 90, and that the pointer of the
instrument shall be directed over a reading of at least 80 or say
0
,

E by N, on the card at this deflection, and that the same angle shall
be used throughout the swing.
(2) To correct the compass:
(a) The Flinders bar and the spheres are first placed in position,
the latter temporarily.
(b) The ship's head must be steadied on the requisite point by
the compass which is being corrected, and then maintained on that
course by other means (steering compass) while the observation is
being made.
(c) To correct Coefficient B. Adjust the scale reading of the
instrument to the mean of the readings on North and South while
heading on one of those points and insert fore and aft magnets,
adjusting them to obtain the 90 deflection with the pointer over E
by N of the card. If the reading on North is the greater, the magnets
must have their red poles forward, and if the lesser then red poles aft.
(d) To correct Coefficient C. Adjust the reading of the scale to the
mean of the readings on East and West while heading on one of
these points and insert athwartship magnets, adjusting them to
obtain the 90 deflection with the pointer over E by N. If the reading
on East is the greater, the magnets must have their red poles to
port, and if the lesser then red poles to starboard.
(e) To correct Coefficient D. Adjust the scale reading of the
instrument to the mean of the above two means and, while heading
East or West, move the spheres to obtain the 90 deflection with
the pointer over E by N. If the mean of North and South readings
is the greater, the spheres must be moved in, and if the lesser then
the spheres must be moved out.
(3) To evaluate ),. Obtain an additional observation with the
compass and bowl landed at a place free from local magnetic fields.
Multiply each of the four readings obtained on board by the cosine
of the respective deviation (if the deviations are large and are
known), mean the results, and divide this by the reading obtained
on shore.
(4) To evaluate the Coefficients B Co and DO:
O
,

(a) Divide 4 by the sum ofthe four readings on North, East, South
 CHAPTER XVII

THE MECHANICAL CORRECTION OF THE

COMPASS

IT will be realised from the theoretical considerations discussed in

the foregoing chapters that certain precautions must be taken if
the compass is to give satisfactory results. These may be enumerated
under the following headings.
The position of the compass. In merchant ships the placing of
the standard compass does not always receive all the consideration
it deserves, but the following rules should, as far as possible, be
adhered to :
(1) The standard compass should be placed so as to obtain a
clear view of as much of the horizon as possible.
(2) It should be in the centre line of the ship and as far away as
possible from large masses of magnetic material, especially those
giving vertical effects, and from movable iron.
(3) No magnetic material should be in any direction nearer than
10 feet (3 m) from the standard compass or 6 feet (2 m) from a
steering compass.
(4) No electrical or electro-magnetic instruments should be near
enough to any compass to have any effect on it.
(5) All electric leads in the vicinity of the compass should be
run so that the supply and return leads of the same circuit are
clipped together with non-magnetic clips, and secured in position
by non-magnetic fastenings.
Note. The Admiralty Compass Department publish a pamphlet
giving the "safe distances" for most instruments and apparatus
fitted in ships. The Ministry of Transport also lay down "safe" and
"conceded" distances for certain equipment. The "safe" distances
should be used whenever possible.
The ship should be swung for deviations of the compass, which
may require adjustment:
(1) After the ship has suffered any severe impact such as collision.
(2) After being struck by lightning.
(3) After any major structural alteration or major repairs.
(4) After loading or discharging by means of electro-magnets.
(5) After lying in one direction for a long period of time.
145
146 MAGNETIC COMPASS DEVIATION AND CORRECTION
(6) If any of the correctors have been moved for reasons other
than adjustment.
(7) At least once a year.
Note. The carriage of cargoes containing magnetic material may
affect deviations during the voyage.
Precautions before swinging for deviations and/or adjustment:
(1) The ship should be upright.
(2) The funnels should be at their sea-going temperature.
(3) All movable iron should be in its sea-going position.
(4) No other ship should be within 3 cables of the ship during
the swing.
(5) The azimuth mirror should be tested and, if necessary,
adjusted.
(6) The lubber point may require checking to make certain it is
in the fore and aft vertical plane through the pivot.
(7) The compass card should be tested for friction by deflecting
the north point about 2 to the right by means of a magnet and
then about 2 to the left. If the card returns to its previous position
of rest, as indicated by the direction of the ship's head, after each
deflection there is no friction.
Precau tions when adjusting:
(1) The permanent magnets must not be placed within a distance
of twice their length from the compass needles, or within a distance
of six times the length of the longest compass needles, which ever
is the greater.
(2) The vertical fore and aft plane passing through the centre
of the compass needle system must pass through the centre of all
athwartship magnets, the centre of the vertical magnet system
(heeling error magn~ts), and the longitudinal axis of the Flinders bar.
(3) The vertical athwartship plane passing through the centre
of the compass needle system must pass through the centre of all
fore and aft magnets, the centre of the vertical magnet system, and
the centre of the spheres.
(4) The horizontal plane passing through the centre of the com-
pass needle system must pass through the centre of the spheres
and a point on the Flinders bar one-twelfth (1/12) of its length from
the upper end.
Note. The distance of the spheres from the compass needle
system must be such that Coefficient D (and E if necessary) is
corrected. Induction in the spheres by the needles is reduced to a
minimum if the needles of the system are short compared with the
distance of the centre of the system from the centre of the spheres.
THE MECHANICAL CORRECTION OF THE COMPASS 147
This is normally the case in the better types of modern compass
cards.
It must also be borne in mind that the proper system of needles
is such that the like poles of symmetrical pairs sub tend an angle
of 60 at the centre, or sextantal deviations may be present.
Order of Correction:
(1) The spheres should be placed to correct for Coefficient D
(and for E if necessary). If the value of the deviation it is required
to correct is known, the distance of the spheres from the compass
may be found from Tables. If not, place the spheres half-way along
the brackets. The spheres should be suitably marked to ensure
that in the event of any further adjustment they are not turned in
azimuth, so that they may retain the same orientation with respect
to the fore and aft line of the ship.
An approximate value for D could be found by observing the
deviations on the four quadrantal points; this necessitates an
additional swing. If the spheres are already in place, by consulting
the records of deviations, it maybe ascertained if ma terial altera tion
in their position is required.
(2) The Flinders bar is then placed if not already in position.
If the ship is new, the position and amount known to be fitted in a
similar type ship should be used. If this is not known, 12 inches
(305 cm) of Flinders bar may be placed on the side of the binnacle
toward the nearer end of the ship; this assumes that the compass
is forward of and above the centre of the ship's superstructure
and hull.
As the Flinders bar acts as a small sphere giving a small + a
rod and a small - e rod effect, this may necessitate the spheres
being moved in about half an inch.
The Flinders bar must be slewed if necessary to correct for any
f rod effect, if presen t.
(3) The heeling error should now be corrected with the aid of
the Heeling Error or Vertical Force Instrument as described on
page 100 et seq.
The ship's head should be placed in an easterly or westerly
direction if there is likelihood of any g rod effect, otherwise the
direction of the ship's head is immaterial. In choosing a place
ashore free from local interference, the instrument should be kept
at least three feet from the ground.
(4) The horizontal permanent magnets are next placed for the
correction of Coefficients Band C. The coefficient having the
greater value should be attended to first.
 148 MAGNETIC COMPASS DEVIATION AND CORRECTION
THE MECHANICAL CORRECTION OF THE COMPASS 149
(a) Coefficient B is usually the greater, in which case the ship's
The deviations of the steering or the after compass are obtained
head should be placed East or West by compass and fore and aft
from comparison with the standard compass.
magnets inserted until there is no deviation showing.
It is possible, after a vessel has been struck by lightning, for
(b) Then place the ship's head North or South by compass and
its compass to become frozen, that is to say, the North point of its
correct Coefficient C by inserting athwartship permanent magnets
until there is no deviation showing. card will point in a certain direction with reference to the fore and
aft line of the ship irrespective of the direction in which the ship
(c) If there was a large deviation on North or South, which ever
is heading. This would indicate that an extremely strong blue
was used, go back to the same heading used for (a) and readjust
pole had developed in that part of the ship due to the electrical
the fore and aft magnets. This is done because a large uncorrected discharge.
Coefficient C would be affecting the directive force at the compass
position when B was first corrected. Before any normal compensation can be made the compass must
be freed by inserting fore and aft and/or athwartship magnets in
(d) Next place the ship on the opposite heading to (a) and halve
the binnacle in such a way that the compass will again respond to
any deviation showing by re-adjusting the fore and aft
magnets. alterations of course. For instance, assume that the North point
of the card tended to point two points on the port bow for all
(e) Now place the ship on the opposite heading to that used in
headings of the ship-place the ship heading about SSW so that
(b), halving the remaining deviation by re-adjusting the athwartship
magnets. the North point of the card is pointing approximately South, then
insert fore and aft magnets, red ends forward, and athwartship
(f) Then place the ship's head on the quadrantal point between
magnets, red ends to port, until the card is able to approximately
the headings used in (d) and (e) and make any necessary final
adjustment to the position of the spheres. reverse its direction. In this particular case the fore and aft com-
ponent should obviously be the stronger of the two.
(g) Finally, swing the ship and obtain the residual deviations
A similar state might be found at the steering compass position
on at least eight, but preferably sixteen, equidistant points. This
of a ship due to bad siting, for instance the binnacle having been
operation should take at least 40 minutes, and the ship should be
placed too close to a bulkhead. The compass must then be freed
steadied on each point; this admits also a more accurate com-
parison with the steering compass being made. as described above before the ordinary correction of the compass
is carried out. It is to be hoped, however, that such a condition
(h) If required, ..1.2 may now be found as described on page 121
. et seq. will seldom be found in ships of the present day.
Note. If bearings of a distant object are used to obtain the
deviations, the effect of parallax must be borne in mind. The
parallax in a 100 metre radius of swing at six miles distance is
about half a degree.
If a shore compass is used to obtain reciprocal bearings, an
efficient means of signalling must be arranged.
If bearings of a heavenly body are used, it is more convenient
to work out and tabulate the magnetic bearings beforehand covering
the period of time of the swing.
When the bearings of two or more known terrestrial objects in
line are used, and the deviations obtained from transits, no particular
precautions are required.
If a gyro compass is used it must be ascertained that it has
no error, but one or other of the methods mentioned above is
preferable.
 INDEX

PAGE PAGE

~ Coefficient, approximate 83 Deflector . 124

A, Coefficient, exact .74, 76, 96 Errors of 131
Abnormal variation 42 Summary of correction by 130
Aclinic line xv, 39 Demagnetisation xv, 25
Agonic line xv, 38 Deviation of compass 53
Ampere 34 analysis of 107
Ampere per metre xv, 34 constant 71, 75
Ampere's rule 32 general expression for 83
Analysis of deviations 107 quadrantal 72, 75
Approximate coefficients 83 semi-circular. 72,76
Dip circle . 39
Ii, Coefficient, approximate 83 magnetic xvi,39
B, Coefficient, exact . . 74, 76, 96 Du Bois equations for magnetic
Binnacle 53 shielding 120
Broadside-on position of magnet
29,46 E, Coefficient, approximate 83
E, Coefficient, exact . 74, 75, 96
C; Coefficient, approximate 83 Earth's magnetism 36
C, Coefficient, exact . . 74, 76, 96 Electro-magnets 34
Circular measure 18 effect of loading and
Coefficients, approximate . 83 discharging by 21
determination of, by analysis Electron theory, flow of current 32
107, 127 Elements, magnetic . 37
determination of, by deflector End-on position of a magnet 28,44
readings 127 Electric wiring in vicinity of
exact 74 compass 145
J, the heeling error 98 Equator, magnetic 37
Lambda 73 Equivalent length of a magnet
Mu 99 xv, 25, 48
order of correction 145 Errors of the deflector 131
Coercive force . xv Exact coefficients 74
Compass, deviation of 53
card 53 F, approximate coefficient 83
Correction of compass, principle Field, magnetic. xvi,22
of 55 due to a current flowing in a
Theory of mechanical 113 wire 31
Cosecant . 17, 18 of a solenoid. 33
Cosine II, 18 strength. xvi,23
Cotangent. 17, 18 Flinder's bar 54, 113, 144
Current, definition of unit. 34 slewing of . 114
Flux, magnetic . xvi,25
D, Coefficient, approximate 83 Force between magnetic poles 22
D, Coefficient, exact . . 74, 75, 97 lines of . 23

163
164 INDEX INDEX 165

PAGE PAGE PAGE PAGE

Frozen compass 149 meridian 38 Rods, the nine . 60 Sub-permanent magnetism xvii,20
moment. xvi,24 effect when the ship heels 87 Sunspot activity 41
G, approximate coefficient 83 permeability . xvi,26 to determine the signs of 62 Susceptibility, magnetic xvi,26
Gaussin error 117 poles xvi,22 Swinging ship for deviations 145
shielding xvi,119 Safe distances 145
H, approximate coefficient 83 susceptibility. xvi,26 Shielding, magnetic xvi, 119 .108, IlIa
Table, analysis.
Hard iron . 19 variation xvi,37 Secant 17 11
Tangent
Harvey Rayne's Heeling Error Magnetisation, intensity of xv, 25 Ship's multiplier 102 124
Thompson deflector .
Corrector 102 Magnetism, induced . xv, 60 Sine 11 Total magnetic force, earth's 40
Heeling error 86 molecular theory of 19 Slewing of Flinder's bar 114 Trigonometrical ratios 11
Coefficient J 98 permanent xvii, 19, 56 of spheres 117
correction of . 100 sub-permanent xvii,20 Soft iron 19
Variation, abnormal . 42
principal causes of. 98 Magnetometer . xvi,47 keepers. 26
magnetic xvi,37
Hysteresis. xv, 117 Moment of inertia xv, 50 Solenoid, magnetic field of 33
Vector 6
Mu, coefficient . 99 Small angles 18,83
Vertical force, earth's 8,40
Induction, magnetic . xvi, 25 Multiplier, ship's . 102 Spheres, soft iron correcting 115
in soft iron correctors 115 instrument 100
slewing of 117
ship's 57, 66
Inertia, moment of 50 theory of use of 136
Neutral axis of a magnet xvii, 22 Vibration experiment 49
effect on deflector readings 135 Steering compass, deviations of 149
Normal deflection 125
Intensity, magnetic 23 Storms, magnetic 42
of magnetisation xv,25 Strength of magnetic field xvi, 23 Weber XVII
Inverse square law xv,23 Order of compass correction 147 49 Wiring, electric, in vicinity of
comparison of
Isoclinal xv, 39 Strength of poles of a magnet 23 compass 145
Isogonal xv, 38 P, force 58
Parallax, effect when swinging
Keepers, soft iron 26 for deviations 148
Kew magnetometer 38 Parallelogram of forces 7
Period of vibration of a magnet 49
Lambda, coefficient 73 Permanent magnetism xvii, 19,56
determination of 121 Permeability, magnetic xvi,26
Least squares principle xv, 107 Pitching error 105
Lightning, effect of ships being Point pole. XVI
struck by 21 Poles, magnetic. 22
Lines of force, magnetic 23 magnetic, earth's 36
Local magnetic disturbance 42 Pole strength, magnetic 23
Position of compass . 145
Magnet, electro- 34 Precautions before swinging 145
permanent 19 Principle of mechanical
Magnetic axis 22 correction of compass 10,55
dip and method of measuring
xvi, 39
Q, force 58
disturbance, local 42
Quadrantal correctors 115
field xvi,22
deviation 72, 75
field, earth's . 36
field strength. xvi,23
foci 37 R, force 58
induction xvi,25 Radians 18
intensity 23 Ratios, trigonometrical 11
latitude . . 170 Retentive error. 118